Investigation of Different Approaches to Real-Time Control of Prosthetic Hands With Electromyography Signals

In this article, we describe a real-time system for prosthetic hands control. The system architecture includes the integration of the electromyographic (EMG) signal acquisition devices, platform for the implementation of the real-time classifier, sensors for the detection of object slip after grasp and the open-source hand prosthesis. Three databases were used to evaluate the implemented classifiers: a database with EMG data from local volunteers and NinaPro DB2 and DB3 databases that include electromyography and accelerometry (ACC) data acquisitions. A Multilayer Perceptron (MLP) classifier was implemented on a platform for rapid prototyping (Raspberry Pi 3 model B+) and generated responses in real-time (11 ms) with average accuracy of 96.30% for 11 hand and wrist gestures/movements.